For many years it has been an aim of scientists in the field of specifically targeted drug therapy to use monoclonal antibodies (MAbs) for the specific delivery of toxic agents to human cancers. Conjugates of tumor-associated MAbs and suitable toxic agents have been developed, but have had mixed success in the therapy of cancer, and virtually no application in other diseases, such as infectious and autoimmune diseases. The toxic agent is most commonly a chemotherapy drug, although particle-emitting radionuclides, or bacterial or plant toxins have also been conjugated to MAbs, especially for the therapy of cancer (Sharkey and Goldenberg, C A Cancer J Clin. 2006 July-August; 56(4):226-243) and, more recently, with radioimmunoconjugates for the preclinical therapy of certain infectious diseases (Dadachova and Casadevall, Q J Nucl Med Mol Imaging 2006; 50(3): 193-204; incorporated herein by reference in its entirety).
The advantages of using MAb-chemotherapy drug conjugates are that (a) the chemotherapy drug itself is structurally well defined; (b) the chemotherapy drug is linked to the MAb protein using very well defined conjugation chemistries, often at specific sites remote from the MAbs antigen binding regions; (c) MAb-chemotherapy drug conjugates can be made more reproducibly than chemical conjugates involving MAbs and bacterial or plant toxins, and as such are more amenable to commercial development and regulatory approval; and (d) the MAb-chemotherapy drug conjugates are orders of magnitude less toxic systemically than radionuclide MAb conjugates.
The present disclosure solves specific problems associated with the preparation of conjugates of the camptothecin (CPT) group of cytotoxic compounds. CPT and its derivatives are a class of potent antitumor agents. Irinotecan (also referred to as CPT-11) and topotecan are CPT analogs that are approved cancer therapeutics (Iyer and Ratain, Cancer Chemother. Phamacol. 42: S31-S43 (1998)). CPTs act by inhibiting topoisomerase I enzyme by stabilizing topoisomerase I-DNA complex (Liu, et al. in The Camptothecins: Unfolding Their Anticancer Potential, Liehr J. G., Giovanella, B. C. and Verschraegen (eds), NY Acad Sci., NY 922:1-10 (2000)).
CPTs present a set of caveats in the preparation of conjugates. One caveat is the insolubility of most CPT derivatives in aqueous buffers. Secondly, CPTs provide specific challenges for structural modification for conjugating to macromolecules. For instance, CPT itself contains only a tertiary hydroxyl group in ring-E. The hydroxyl functional group in the case of CPT must be coupled to a linker suitable for subsequent protein conjugation; and in potent CPT derivatives, such as SN-38, the active metabolite of the chemotherapeutic CPT-11, and other C-10-hydroxyl-containing derivatives such as topotecan and 10-hydroxy-CPT, the presence of phenolic hydroxyl at C-10 position complicates the necessary C-20-hydroxyl derivatization. Thirdly the lability of the δ-lactone moiety of the E-ring of their structures, under physiological conditions, results in greatly reduced antitumor potency of these products. Therefore, the conjugation protocol is performed such that it is carried out at a pH of 7 or lower to avoid the lactone ring opening. Typically conjugation of a bifunctional CPT possessing an amine-reactive group such as an active ester would require a pH of 8 or greater. Fourth, an intracellularly-cleavable moiety is to be incorporated in the linker/spacer connecting the CPTs and the antibodies or other binding moieties.
The problem of δ-lactone opening under physiological conditions has been previously addressed. One approach has been to acylate the C-20 hydroxyl group with an amino acid, and couple the α-amino group of the amino acid to poly-L-glutamic acid (Singer et al. in The Camptothecins: Unfolding Their Anticancer Potential, Liehr J. G., Giovanella, B. C. and Verschraegen (eds), NY Acad Sci., NY 922:136-150 (2000)). This approach relies on the passive diffusion of a polymeric molecule into tumor sites. This glycine conjugation has also been reported as a method of making water-soluble derivative of CPT (Vishnuvajjala et al., U.S. Pat. No. 4,943,579) and in the preparation of a PEG-derivatization of CPT (Greenwald, et al. J. Med. Chem. 39: 1938-1940 (1996). In the latter case, the approach has been devised in the context of developing water-soluble and long acting forms of CPT, whereby CPT's in vivo half-life is enhanced, and the drug is gradually released from its conjugate while in circulation in vivo.
The present invention discloses methods for preparing conjugates of CPTs, of 10-hydroxy derivatives such as SN-38 in particular, taking into consideration the four caveats described above and the synthetic challenges. SN-38 is the active drug form of the approved cancer drug CPT-11, which is a prodrug. Vast clinical data are available concerning CPT-11 pharmacology and of its in vivo conversion to SN-38 (Iyer and Ratain, supra; Mathijssen et al., Clin Cancer Res. 7:2182-2194 (2002); Rivory, Ann NY Acad Sci. 922:205-215, 2000)). The active form SN-38 is about 2 to 3 orders of magnitude more potent than CPT-11.
Early work on protein-drug conjugates indicated that a drug ideally needed to be released in its original form, once it had been internalized into a target cell, for the protein-chemotherapy drug conjugate to be a useful therapeutic. Trouet et al. (Proc. Natl. Acad. Sci. USA 79:626-629 (1982)) showed the advantage of using specific peptide linkers, between the drug and the targeting moiety, which are cleaved lysosomally to liberate the intact drug. Work during the 1980's and early 1990's focused further on the nature of the chemical linker between the chemotherapeutic drug and the MAb. Notably, MAb-chemotherapy drug conjugates prepared using mild acid-cleavable linkers were developed, based on the observation that the pH inside tumors was often lower than normal physiological pH. In this respect, superior results were found by incorporating a hydrazone as a cleavable unit, and attaching DOX to a MAb via a thioether group, (Willner et al., U.S. Pat. No. 5,708,146; Trail et al. (Science 261:212-215 (1993)).
This approach showed that MAb-doxorubicin (DOX) conjugates, prepared with appropriate linkers, could be used to cure mice bearing a variety of human tumor xenografts, in preclinical studies. The first approved MAb-drug conjugate, Gemtuzumab Ozogamicin, incorporates a similar acid-labile hydrazone bond between an anti-CD33 antibody, humanized P67.6, and a potent calicheamicin derivative. Sievers et al., J Clin Oncol. 19:3244-3254 (2001); Hamann et al., Bioconjugate Chem. 13: 47-58 (2002). In some cases, the MAb-chemotherapy drug conjugates were made with reductively labile hindered disulfide bonds between the chemotherapy drugs and the MAb (Liu et al., Proc Natl Acad Sci USA 93: 8618-8623 (1996)). Yet another cleavable linker involves a cathepsin B-labile dipeptide spacers, such as Phe-Lys or Val-Cit, similar to the lysosomally labile peptide spacers of Trouet et al. containing from one to four amino acids, which additionally incorporated a collapsible spacer between the drug and the dipeptide (Dubowchik, et al., Bioconjugate Chem. 13:855-869 (2002); Firestone et al., U.S. Pat. No. 6,214,345 B1; Doronina et al., Nat. Biotechnol. 21: 778-784 (2003)). The latter approaches were also utilized in the preparation of an immunoconjugate of camptothecin (Walker et al., Bioorg Med Chem Lett. 12:217-219 (2002)). Another cleavable moiety that has been explored is an ester linkage incorporated into the linker between the antibody and the chemotherapy drug. Gillimard and Saragovi have found that when an ester of paclitaxel was conjugated to anti-rat p75 MAb, MC192, or anti-human TrkA MAb, 5C3, the conjugate was found to exhibit target-specific toxicity. Gillimard and Saragovi, Cancer Res. 61:694-699 (2001).
While the importance of cleavable linker in the design of binding moiety-drug conjugates cannot be overstated, it is also important to focus on how the linker design impacts the overall preparation of specific CPT-binding moiety conjugates. The present invention solves the problem associated with the preparation of the bifunctional drug-linker molecule, wherein the said drug may also contain more than one reactive group for derivatization, such as the potent SN-38 analog, for instance, in the design of conjugates. SN-38, a clinically important active drug form of the cancer drug CPT-11, but 100-1000-times more potent than CPT-11, is not useable systemically because of insolubility. The present invention solves this problem by conjugating it to a targeting moiety in ways that also address other challenges of using a CPT, while concurrently improving the therapeutic index of this clinically important potent drug by using disease-specific antibodies.
The conjugates of the instant invention possess greater efficacy, in many cases, than unconjugated or “naked” antibodies or antibody fragments, although such unconjugated targeting molecules have been of use in specific situations. In cancer, for example, naked antibodies have come to play a role in the treatment of lymphomas (CAMPATH® and RITUXAN®), colorectal and other cancers (ERBITUX® and AVASTIN®), breast cancer (HERECEPTIN®), as well as a large number now in clinical development (e.g., epratuzumab). In most of these cases, clinical use has involved combining these naked, or unconjugated, antibodies with other therapies, such as chemotherapy or radiation therapy.
A variety of antibodies are also in use for the treatment of autoimmune and other immune dysregulatory diseases, such as tumor necrosis factor (TNF) and B-cell (RITUXAN®) antibodies in arthritis, and are being investigated in other such diseases, such as the B-cell antibodies, RITUXAN® and epratuzumab, in systemic lupus erythematosus and Sjögren's syndrome, as well as juvenile diabetes and multiple sclerosis. Naked antibodies are also being studied in sepsis and septic shock, Alzheimer's disease, and infectious diseases. The development of anti-infective monoclonal antibodies has been reviewed recently by Reichert and Dewitz (Nat Rev Drug Discovery 2006; 5:191-195), incorporated herein by reference, which summarizes the priority pathogens against which naked antibody therapy has been pursued, resulting in only 2 pathogens against which antibodies are either in Phase III clinical trials or are being marketed (respiratory syncytial virus and methicillin-resistant Staphylococcus aureus), with 25 others in clinical studies and 20 discontinued during clinical study.
Thus, there is a need to develop more potent anti-pathogen antibodies and other binding moieties. Such antibody-mediated therapeutics can be developed for the treatment of many different pathogens, including bacteria, fungi, viruses, and parasites, either as naked (unconjugated), radiolabeled, or drug/toxin conjugates. In the case of delivering drug/toxin or radionuclide conjugates, this can be accomplished by direct antibody conjugation or by indirect methods, referred to as pretargeting, where a bispecific antibody is used to target to the lesion, while the therapeutic agent is secondarily targeted by binding to one of the arms of the bispecific antibody that has localized at the site of the pathogen or of the cancer or whatever lesion is being treated (discussed by Goldenberg et al., J Clin Oncol. 2006 February 10; 24(5):823-34; and Goldenberg et al., J Nucl Med. 2008 January; 49(1):158-63, both incorporated in their entirety herein by reference).